[log in to unmask] wrote:
>
> Could I please get some feedback as to the definitions and difference between
> characteristic versus differential impedance? I have customers that ask for
> both but can't explain what they are actually looking for. Any help would be
> appreciated.
> Thanks,
> Scott
This may be long so I apoligize ...
This is a really good question. I'm going to
stick my neck out to clarify or to have clarified
what I understand about it. And I would invite any
comments as well as flames. <I'm wearing my flame suit>
"Resistance" is the property whereby electrical energy gets
converted into heat. It represents a complete loss. In
other words, you put current through a resistor and it
will heat up. BUT, try heating the resistor with a heat
gun and you won't get current.
"Inductance" is the property whereby electrical energy gets
converted to a magnetic field. "Capacitance" is the property
whereby electrical energy gets converted to an electrical field.
These act differently than resistance. In other words, under
ideal conditions, the energy that you put IN to inductance or
capacitance will give the same energy OUT. Inductive or
capacitive "reactance" is similar to resistance in that they
oppose electrical energy be it current or voltage respectively.
But that's about where all the similarity stops.
"Impedance" is a combination of resistance, capacitance
and inductance. So now you start getting some loss
depending upon alot of factors.
"Characteristic Impedance" should not be thought of as a loss.
Say, for instance, you terminate a 50 ohm trace with a 50 ohm
resistor. Put a signal down that trace. Measure the current
in that trace. Then, you calculate power as P = (R)*(I^2),
plug in R the characteristic impedence of a trace, plug in I
the current of the signal down that trace, to calculate the
power dissipated by the trace as you would in the terminating
resistor - it would be wrong. The characteristic impedence
of the trace is not acting the same as the resistor.
It's more like a merit of quality for signal concerns when
matching things such as properties of a trace to properties
of devices that connect to that trace. And it has little to do
with length in the theoretical world. It is based on primarily
two things: 1) the fact that there is some known datum for ground,
in fact a continuous ground, and 2) geometry of the trace.
Phew ... Ok, now for something slightly different.
Let's talk about two traces over a common ground plane.
Trace 1 Trace 2
______ ______
______________________
Ground Plane
"Common Mode" Impedance is where the characteristic impedance
of the traces is considered in relation to the ground plane.
"Differential Mode" Impedance is where the characteristic
impedance factors in the other trace as well. It looks something
like this ...
Trace 1 Zd Trace 2
______ /\/\/\ ______
\ \
Zc1 / / Zc2
\ \
______________________
Ground Plane
"Even Mode" Impedance is when signals on both traces are
in exact step with one another. So, their effects (side to
side) to each other differentially cancel out and thus it's
essentially a common mode effect - ground is the only thing
that comes into play.
"Odd Mode" Impedence is when signals on both traces are
out of step with one another. So, their effects to each
other do NOT cancel one another. So, their effects to each
other include differential mode as well.
Ok, that's enough ...
################################################################
TechNet E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c
################################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body:
To subscribe: SUBSCRIBE TechNet <your full name>
To unsubscribe: SIGNOFF TechNet
################################################################
Please visit IPC's web site (http://www.ipc.org) "On-Line Services" section for additional information.
For technical support contact Hugo Scaramuzza at [log in to unmask] or 847-509-9700 ext.312
################################################################
|
